Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

Voice Output Timing Control Method for a Voice Output Device

IP.com Disclosure Number: IPCOM000123791D
Original Publication Date: 1999-May-01
Included in the Prior Art Database: 2005-Apr-05
Document File: 1 page(s) / 42K

Publishing Venue

IBM

Related People

Sakamoto, M: AUTHOR [+3]

Abstract

Disclosed is a method for automatically delaying or repeating voice output, so that a user will not fail to hear it in noisy surroundings. Voice output systems are widely used to provide guidance in vending machines and announcement at train station. In these applications, the surrounding noise level is constant. However, since the advent of mobile devices with voice output functions, the surrounding noise level can not be fixed. For mobile use, a method for controlling the voice output is necessary, to make it easy to hear the voice in noisy surroundings.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 85% of the total text.

Voice Output Timing Control Method for a Voice Output Device

   Disclosed is a method for automatically delaying or repeating
voice output, so that a user will not fail to hear it in noisy
surroundings.  Voice output systems are widely used to provide
guidance in vending machines and announcement at train station.  In
these applications, the surrounding noise level is constant.
However, since the advent of mobile devices with voice output
functions, the surrounding noise level can not be fixed.  For mobile
use, a method for controlling the voice output is necessary, to make
it easy to hear the voice in noisy surroundings.

   In our method, the surrounding noise level is always
estimated by using a running histogram of the surrounding noise
level.  When the surrounding noise level exceeds a threshold, this
method controls the voice output as follows: If the voice output has
not started, it is delayed until the surround noise level drops.  If
the voice output has started, the current word or sentence is
repeated.

   To avoid the influence of disturbing noises, the voice
output is started 50 milliseconds after the surrounding sound level
dropped.  The influence of a disturbing noise at the time of its
onset is extremely strong.  For this reason, the user sometimes fails
hear the voice when it overlaps with the onset of a disturbing noise.
Therefore, the word or the sentence needs to be output again.

   The figure shows the voice output system with the output
cont...